Long run recursive VAR models and QR decompositions
Long run recursive VAR models and QR decompositions
Long-run recursive identification schemes are very popular in the structural VAR literature. This note suggests a two-step procedure based on QR decompositions as a solution algorithm for this type of identification problem. Our procedure will always deliver the exact solution and it is much easier to implement than a Newton-type iteration algorithm. It may therefore be very useful whenever quick and precise solutions of a long-run recursive schemes are required, e.g. in bootstrapping confidence intervals for impulse responses
University of Southampton
Hoffmann, Mathias
b40fe4f6-c63c-472c-ab67-52702d1f9b00
2000
Hoffmann, Mathias
b40fe4f6-c63c-472c-ab67-52702d1f9b00
Hoffmann, Mathias
(2000)
Long run recursive VAR models and QR decompositions
(Discussion Papers in Economics and Econometrics, 15)
Southampton, UK.
University of Southampton
10pp.
Record type:
Monograph
(Discussion Paper)
Abstract
Long-run recursive identification schemes are very popular in the structural VAR literature. This note suggests a two-step procedure based on QR decompositions as a solution algorithm for this type of identification problem. Our procedure will always deliver the exact solution and it is much easier to implement than a Newton-type iteration algorithm. It may therefore be very useful whenever quick and precise solutions of a long-run recursive schemes are required, e.g. in bootstrapping confidence intervals for impulse responses
More information
Published date: 2000
Identifiers
Local EPrints ID: 33116
URI: http://eprints.soton.ac.uk/id/eprint/33116
PURE UUID: 42a8f170-29d2-4b70-9a9e-775ed29b4718
Catalogue record
Date deposited: 19 Jul 2006
Last modified: 15 Mar 2024 07:42
Export record
Contributors
Author:
Mathias Hoffmann
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics